Hello, Thanks for asking!
Have you checked for combustion gases in the coolant- using a "block check" tool ?
The 3.5 are not known for loosing head gaskets or cracking heads- but it could .
Other problems we have seen are bad water pumps. Check for flow at the overflow bottle . There is a line that purges air into the bottle and one that recirculates it back to the engine. So you should be able to see the flow thru the bottle.
You should not be boiling over at those temps.A bad cap would not seal and allow coolant to overfill the overflow and reduce boiling point- But a good coolant mix will not boil at that temp- I still think you may have compression gases leaking in. Hope This helps. If you need more ideas or tip. Please reply. Thanks Scott.
No luck. Replaced water pump (it was fine). Smog machine does not detect any combustion gases in coolant. Could this be a computer reflash issue? The computer is seeing the correct voltage from the temp sensor (tried three different - all read the same.) Can you tell me the voltage to indicated temp? For instance, 1 volt out of the temp sender to the computer should indicate what temp F? At what voltage should the fans come on? The cooling system physically seems to be in order....it just can't get to 225F in the computer to get the fans on, when the actual coolant temp is 235F-240F. If you can provide the correct voltage I can check.
Steve - the temp sensor voltage and water temp cross around 200-215 deg = 2.00 - 2.15 volts As the temp goes up the volts go down.
There is no reflash for this concern. Most - if not all flashes deal with reporting false check engine lights and tweaking air fuel mixtures.
You should see about 4.98 volt measured or read on the scanner with the sensor unplugged. Looking at the wiring diagram - the sensor -coolant - is provided its own 5 volt signal and shares a common sensor ground. Its possible that the pcm is not able to supply the proper voltage and current to provide the correct voltage drop that is needed to signal a fan request- very rare . That would set a fault over time.
If your scanner can see fan request and also can actuate the fan relay - try actuating the cooling fans . turn the a/c on and monitor the a/c pressure and request for cooing fan from the pcm . I checked for the temp vs volts table- but if think thats listed in one of the student training books - that we all loose or toss out! .
You should have 50% fan speed at 220 deg and shuts off at 214 deg- The fan relay is electronic and controlled by a pulse width 12 volt signal from the PCM .
Let me know if you can control the fan relay with your scan tool. and if the A/C will turn the relay on- You may have a failing relay .
The radiator may be plugged up and not allowing coolant flow , and this would cause the
overheat condition also- I am not sure if you can get a look at the inlet side of the radiator to inspect for plugging.
Seems you have checked every angle . You would have to use a cooling system flow tester connected to the rad outlet side. You should also be able to feel or shoot a temp readings across the radiator when the t- stat opens. Cool areas will be blockage .
Hope this helps . Please reply if you need more help or if clarifications . Thanks Scott
I have the 5 volt signal at the connector. But when the vehicle heats up, 1 volt equals 190. The sensor is producing 1 volt (with a volt meter) and the computer sees 1 volt (scanner) and the temp shows about 190. Unfortunately, the actual temp at the housing is already approaching 220F. The thermostat is working properly. A/C activation turns fan on and actual water temp drops immediately -- radiator is flowing well. If I put the sensor in hot water and heat it must drop to .64V before the fans come on at 225F on the scanner. Relay is new and working as should. The issue must be between the sensor and ECM...somehow not enough/actual temp is being reported.?
Ok , since the coolant temp sensor shares a common sensor ground - and has its own dedicated pin on the computer - lets bypass the all the engine wiring and connector .
I have a jumper cable that I use to bypass the wire harness. Since you only have two wires at the sensor its easy to jump. run a new coolant sensor signal wire . From pcm connector 2 - pin 20? - wire diagram I have does not zoom.! wire color violet with orange stripe / to the coolant sensor.
make sure you cut the wire at the pcm and the sensor- that way is any voltage is present on that wire-it will be completely by passed. For now lets assume the signal ground wire splice at S106 is good. See if that straightens it out.
But you said that heating the sensor to 220 will turn the fans on. Well that would tend to indicate that the sensor and wiring are good--- So that would tend to indicate that the sensor is not "seeing " the real temp. - Could the t-stat that you replaced be stuck closed? - check for a HOT radiator hose and coolant flow at 195-197 deg.Could there be an obstruction at the coolant sensor passage- blow it out with air----? .
Sorry I am thinking out loud- Add a new wire to sensor signal first - then splice a signal ground in at pcm connector 2 = pin 27-- this wire should be soldered into the existing wire at terminal 27 of the pcm C2 and cut at the sensor. Thanks about all that left is the wiring. The factory manual shows no connections between the sensor signal and PCM. But there a at least 2 splices and one connector on the sensor ground circuit.
Thanks for the tough one--- please let me know how it works out , or if you need more info or ideas. Thanks Scott=-
P.S. -- make sure you have good computer grounds at C1 pins 9 and 18. You may want to add a new ground at those wires.
Wow! What an ass-kicker! Had to give you the final. Did like you said....cut the coolant temp wires at the computer and at the sensor plug; ran two new wires, soldered and sano. Also piggyback grounded the two grounds you mentioned in the C1 plug directly to the battery. Won't start. Throws 8 codes for high circuits everywhere (IAT, CPS, CKS, Bat Temp, etc..) So, for ducks, I check the two empty wires that used to go to the coolant temp sensor (the ones I cut at the sensor end). Key on, the blu/grn that should be ground for the sensor has 5 volts on it! WTF? So I aligator clip it to the sensor ground that is now wired directly from the sensor to the ecm --- starts right up. So, obviously this ground is tied into all sensor grounds somehow. Shut it off, key on the alligator clipped wire has 1/2 volt on it.....(and open it has 5 volts!) If I watch the scanner, removing the clip lowers my temp sensor voltage .2 and raises my water temp 20F. BAM! There is the missing 20 degrees I needed to turn the fans on! So, my final two questions for you.....any thoughts on where that voltage might be sneaking in on the ground? If not, is there any issue with leaving my direct wire from the ecm ground to the sensor, tied into the original harness again AND piggybacking a ground on top direct to the battery? Seems to like it. Any issues it may cause with the ecm?
Thanks a lot for working me thru this one!
Good- I would look closely at the battery- check for signs of past acid leakage- even a small drop can degrade the insulation on a wire and allow voltage to "seep " .
You have to keep that sensor ground wire connected to provide a circuit for all the other sensors.
The sensor ground and engine ground are not exactly the same thing on many circuits, There are filters and diagnostic voltages on some sensor grounds that can be defeated by a straight engine ground- The most common problem is battery acid - the second is the a/c pressure transducer .
Just make sure that if you add a ground to the sensor ground - that all circuits seem normal- and make sure you tell the customer that a voltage leak into the sensor ground circuit was found- but not pinpointed- and that this could be related to a oil soaked wire, acid damage, wire chaffing , or an internal computer circuit failure - time will tell if its fixed or not,. Just make sure you cover all possible causes- that proves that you know the exact problem - but have not spent the money to find it- but was able to isolate it and repair it -- for now-- . Never be worried about explaining what was found and how it was fixed-Its easier to do this than to have to work for free the next time it comes in- we did not cause the problem- only defined it and tried to override it with out throwing parts at it.
Just plainly explain the problem found - the method to repair and mention that further problems may appear. Seems to have either a wiring harness concern or a computer glitch. Hope this helps, Please reply if you need more help of info- Regards Scott.